| Unmaned/intelligent working face mining technology is an effective and safe way of exploitation,which can effectively reduce the accidents and ensure the safe mining.Coal science and technology 13 th Five-Year clearly pointed out the need of implementation the strategy of innovation driven development,resolving the key technology of intelligent mine and brining the intelligent control technology into the coal production process.In order to achieve this goal,the intelligent identification of coal rock interface should be solved at first which is the key problem restricting the intelligent coal mining equipment.Coal rock interface refers to the interface between coal layer and rock layer.The mining machine should cut along with the coal rock interface and adjust the height of cutting drum immediately to avoid over/under cutting when the coal rock interface changes.The topic has been a hot research at home and abroad.Though many scholars have put forward many representative research methds,each method just discussed the feasibility of coal rock interface recongnition and a final solution has not been formed.In consideration of the electromagnetic wave propagation in the coal seam is affected by the electromagnetc parameters of coal and rock.Whil the electromagnetc parameters of coal and rock vary with frequency.It's possible to avoid the situation that the electromagnetc parameters of coal and rock are close with each other by selectiong an appropriate frequency,which can not be achieved by other methods.In addition,when the frequency of electromagnetic wave eaches the terahertz frequency band,the physical information of coal and rock ban be obtained by studying the electromagnetic wave echo.Therefor,the study of coal rock interface and coal rock medium have been done by using electromagnetic wave technology1)In microwave band,the coal rock interface was set as research object.The electromagnetic prospecting model of coal rock interface was established which was related to the electromagnetic parameters of coal and rock,and geological environment parameters.The identification method and applicability of coal rock interface was studied through this model.The feasibility of GPR technology was analyzed from the aspects of the electrical difference between coal and rock medium,and the distance between the radar system and coal rock interface target.The forward process of electromagnetc wave in coal/rock layer was simulated.The propagation characteristics of electromagnetic wave in complex coal seam were studied,and the applicability of electromagnetic detection under different test frequencies(microwave frequency),various coal rock combinations and simple/complex geological conditions were studied.The results show that most coal rock interface within the distance of 0.3~8 m could be detected by using the LTD-2100 GPR.In order to study the acquisition method of coal rock interface underground,and verify the simulation results through underground case studies,the underground experiments were carried out by using LTD-2100 GPR and different radar antennas.It's found that the experimental data are in good agreement with the theoretical data which means most coal rock interface could be detected by using GPR.Both forward detection and real-time detection which means the radar antenna moves with the shearer are feasible.The better interpretation of the radar image and more accurate information of coal seam thickness could be obtained by combining the geological data to study the characteristics of coal rock interface in radar profile.2)The characteristics of different kinds of coal and rock terahertz signals were extracted.The change rules of absorption,refraction and dielectric properties of coal and rock were revealed based on material composition analysis and Debye/Lorentz model.The feasibility of coal and rock identification in the terahertz frequency band was verified.The typical terahertz signal of coal rock medium was obtained using THzTDS.It's necessary to extract and study the characteristic information of coal rock medium due to the fact that there is no characteristic absorption peak of these medium in terahertz domain.The characteristics and physical properties of coal and rock terahertz signal were studied from time domain,frequency domain and time-frequency domain.The optical sonstant as well as the dielectric properties and loss characteristics of the coal and rock medium in terahertz frequency were obtained by the relationship among the optical constants,conductivity and dielectric constant.Compared the electrical parameters of coal and rock medium at 0.5 THz with 100 MHz,it's found that the dielectric constant of coal and rock decreases when the frequency reaches terahertz band.However,the dielectric constant of rock changes more greatly than coal which means the reflection characteristics of coal wall doesn't change much,while the rock wall changes greatly.In addition,the conductivity of coal and rock increases about 7 orders of magnitude,which means the strong attenuation of coal and rock medium in THz band.A method for studying the dielectric properties of coal and rock in THz band based on Debye/Lorentz model was presented by researching the relationship between frequency and dielectric constant of coal and rock.The results show that the experimental data and theoretical data are in good agreement under three order model,and the root mean square error is less than 0.01 and the average correlation coefficient is up to 0.96.The feasibility of identifying the coal and rock interface in the terahertz frequency band was preliminarily verified by using the difference of coal and rock physical characteristics,as well as the difference of terahertz signal in the time domain,frequency domain and time-frequency domain.What's more,the effects of different coal types,degree of metamorphism,ash content,volatile matter on the physical properties of coal in THz band were studied.It's pointed out that the degree of metamorphism and ash content of coal are the main factors affecting the physical properties of coal in THz band.3)In THz band,a novel method of coal rock identification based on THz-PCASVM was proposed when the character of coal and rock were similar.The principal component analysis was used to reduce the characteristic of absorption data and refraction data of 100 groups coal and rock,and the results of PCA were analyzed.The results show that the contribution of the first three principal components of the absorption spectrum and the refractive index spectrum is more than 99%,which greatly decrease the data dimensionality.Additionally,in the two dimension PCA score of coal rock absorption spectrum,the coal rock medium has not been completely separated,but all kinds of coal were gathered together.While in the two dimension PCA score of coal rock refractive index spectrum,the coal and rock medium are separated completely,but the principal component score of each coal species shows a trend of divergence.THzPCA-SVM method which was proposed for identification of coal and rock by combining the spectral technique and mathematical method.The effects of different kernel functions and support vector machine parameters on the classification results were analyzed.The results showed that after PCA pretreatment both the absorption spectra data sets and the refractive index data sets can be used to construct an accurate classification model and achieve 100% accurate classification results.Even in the case that the coal and rock properties are similar and have no absorption peak,the THz-PCASVM classification method can quickly,efficiently and accurately identify different coal and rock medium.This means the method could be used to describe the cutting state of shearer drum(cutting coal / cutting rock).In addition,the method can also be used as a fast and nondestructive method for distinguishing different kinds of coal,which provides a new idea and solution for the coal quality detection. |
PDF Full Text Request